Computers in Human Behavior 25 (2009) 929939

Contents lists available at ScienceDirect

Computers in Human Behavior

journal homepage: www.elsevier.com/locate/comphumbeh

Fostering the determinants of knowledge sharing in professional

virtual communities
Ming-Ji James Lin a, Shiu-Wan Hung a, Chih-Jou Chen a,b,*
a
Department of Business Administration, National Central University, No. 300, Jhongda Rd., Jongli City, Tao-Yuan 32001, Taiwan, ROC
b
Department of Information Management, National Penghu University, No. 300, Liu-Ho Rd., Makung City 880, Penghu Hsien, Taiwan, ROC

a r t i c l e i n f o a b s t r a c t

Article history: Professional virtual communities (PVCs), which are formed on the Internet, are expected to serve the
Available online 22 April 2009 needs of members for communication, information, and knowledge sharing. The executives of organiza-
tions should consider PVCs as a new innovation or knowledge pool since members share knowledge.
Keywords: However, many PVCs have failed due to members low willingness to share knowledge with other mem-
Knowledge sharing bers. Thus, there is a need to understand and foster the determinants of members knowledge sharing
Virtual communities behavior in PVCs. This study develops an integrated model designed to investigate and explain the rela-
Community loyalty
tionships between contextual factors, personal perceptions of knowledge sharing, knowledge sharing
Norm of reciprocity
Trust
behavior, and community loyalty. Empirical data was collected from three PVCs and tested using struc-
tural equation modeling (SEM) to verify the t of the hypothetical model. The results show that trust sig-
nicantly inuences knowledge sharing self-efcacy, perceived relative advantage and perceived
compatibility, which in turn positively affect knowledge sharing behavior. Furthermore, the study nds
that the norm of reciprocity does not signicantly affect knowledge sharing behavior. The results of the
study can be used to identify the motivation underlying individuals knowledge sharing behavior in PVCs.
By investigating the impacts of contextual factors and personal perceptions on knowledge sharing behav-
ior, the integrated model better explains behavior than other proposed models. This study might help
executives of virtual communities and organizations to manage and promote these determinants of
knowledge sharing to stimulate members willingness to share knowledge and enhance their virtual com-
munity loyalty. As only little empirical research has been conducted on the impact of knowledge sharing
self-efcacy, perceived relative advantage, and perceived compatibility on the individuals knowledge
sharing behavior in PVCs, the empirical evidence reported here makes a valuable contribution in this
highly important area.
2009 Elsevier Ltd. All rights reserved.

1. Introduction
The internet has led to a proliferation of virtual communities
(VCs) all over the world (Fernback, 1999; Hiltz & Wellman,
1997). Exchanging information and knowledge inside VCs has dra-
matically changed our lives. To be successful in todays competitive
workplace, more and more individuals proactively take part in dif-
ferent kinds of VC, especially in professional VCs (PVCs) for knowl-
edge workers that enable them to seek, collect, or even contribute
knowledge to improve their capabilities, to absorb advanced in-
sights, and to resolve problems at work. Many organizations have
also recognized PVCs as valuable systems for knowledge manage-
ment and have begun to support the development and growth of
* Corresponding author. Address: Department of Information Management,
National Penghu University, No. 300, Liu-Ho Rd., Makung City 880, Penghu Hsien,
Taiwan, ROC. Tel.: +886 6 9264115x5622; fax: +886 6 9277401.
E-mail addresses: jameslin@cc.ncu.edu.tw (M.-J.J. Lin), shiuwan@mgt.ncu.edu.tw
(S.-W. Hung), benson@npu.edu.tw, 954401012@cc.ncu.edu.tw (C.-J. Chen).
PVCs to meet their business needs and objectives (Gongla & Rizz-
uto, 2001). The executives of organizations should consider PVCs
as a new innovation or knowledge pool since members share
knowledge (Nambisan & Sawhney, 2007).
Over the past decade, a number of researchers have suggested
that VCs (Preece, 2000; Rothaermel & Sugiyama, 2001) and
knowledge sharing behaviors are inuential to knowledge man-
agement success (Chowdhury, 2005; Kankanhalli, Tan, & Wei,
2005; Wasko & Faraj, 2005; Williams, 2001). This importance
has led to the investigation of knowledge sharing in VCs by some
scholars in an effort to determine what factors are signicant to
knowledge sharing and knowledge management success (Chiu,
Hsu, & Wang, 2006; Hsu, Ju, Yen, & Chang, 2007; Koh & Kim,
2004; Wasko & Faraj, 2005). Most previous studies have focused
on either contextual factors and knowledge sharing (Bock &
Kim, 2002; Bock, Zmud, Kim, & Lee, 2005; Kankanhalli et al.,
2005; Purvis, Sambamurthy, & Zmud, 2001; Wasko & Faraj,
2005) or on personal factors and knowledge sharing (Bock &
Kim, 2002; Chiu et al., 2006; Hsu et al., 2007; Kankanhalli

0747-5632/$ - see front matter 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.chb.2009.03.008
930 M.-J.J. Lin et al. / Computers in Human Behavior 25 (2009) 929939
et al., 2005; Wasko & Faraj, 2005). In this study, we propose an
integrated framework to develop a more comprehensive perspec-
tive of the relationships between contextual factors, personal per-
ceptions of knowledge sharing, knowledge sharing behavior, and
community loyalty, and bring it up-to-date with empirical data
from three PVCs.
We investigate how the contextual factors (norm of reciprocity
and trust) and personal perceptions of knowledge sharing (knowl-
edge sharing self-efcacy, perceived relative advantage, and per-
ceived compatibility) can inuence PVC members willingness to
share knowledge with other members and their loyalty to their
communities. To get a better understanding of knowledge sharing
in PVCs, we examine the relationships between the following four
pairs of concepts:
 Contextual factors and knowledge sharing behavior.
 Contextual factors and personal perceptions of knowledge
sharing.
 Personal perceptions of knowledge sharing and knowledge shar-
ing behavior.
 Knowledge sharing behavior and community loyalty.
We performed an on-line survey in three PVCs in Taiwan. The
empirical data used in the study comprises 350 members, includ-
ing IS (information science) engineers, programmers, managers,
researchers, teachers, students, and other knowledge workers.
The research model and hypothesized relationships are empirically
tested using the structural equation modeling (SEM) approach,
supported by LISREL 8.7 software.
Knowledge-based VCs or organizations need to increase the
quality and quantity of new knowledge more rapidly to satisfy
the expanding requirements of members. This research has been
pursued to assist executives of VCs or organizations to solve
some of the difculties that occur in knowledge management
(KM). A universally accepted denition of KM does not yet exist.
Many denitions of KM have been proposed in the literature
(Nonaka, 1991; Petrash, 1996; Wiig, 1997). Our comprehensive
denition of KM is based on Corso, Martini, Paolucci, and Pelleg-
rini (2003). KM in this context is the combination of manage-
ment systems, organizational mechanisms, information, and
communication technologies through which an organization fos-
ters and focuses individual and group behaviors in terms of
assimilation and generation, transfer and sharing, capitalization
and reuse of knowledge, in tacit or explicit form, that is useful
to the organization.
Even with the existence of information systems, knowledge
sharing is a difcult challenge for organizations (Argote, Ingram,
Levin, & Moreland, 2000; Bakker, Leenders, Garray, Kratzer, &
Van Engelen, 2006; Szulanski, 1996). Researchers have noted that
knowledge management often fails in fostering knowledge shar-
ing efforts due to its neglect of the willingness of knowledge
sharing and the knowledge required for successful knowledge
sharing; the omission of important enhancing activities from their
knowledge sharing mechanisms; and their often ineffective and
inefcient performance of knowledge sharing tasks (Kankanhalli
et al., 2005; Pfeffer & Sutton, 1999). In addition to these difcul-
ties, the investigation of knowledge sharing relationships with
personal perceptions of knowledge sharing such as knowledge
sharing self-efcacy, perceived relative advantage, and perceived
compatibility have not been closely scrutinized to ascertain
members involvement with and contribution to knowledge shar-
ing in VCs. The inuence of contextual factors and personal
perceptions of knowledge sharing on knowledge sharing behavior
and community loyalty is a critical area on which very few stud-
ies have been performed. We attempt to illuminate some of KM
areas.
2. Theoretical background
2.1. Knowledge sharing and inuencing factors
Knowledge sharing is the behavior of an individual dispersing
his or her obtained knowledge and information to other colleagues
within an organization (Ryu, Ho, & Han, 2003). Knowledge sharing
involves a process of communication whereby two or more parties
are involved in the transfer of knowledge. Hence, knowledge shar-
ing is dened as a process of communication between two or more
participants involving the provision and acquisition of knowledge
(Usoro, Sharratt, Tsui, & Shekhar, 2007). Recently, researchers have
highlighted the various factors that affect an individuals willing-
ness to share knowledge, such as information and communication
technologies, costs and benets, incentive systems, extrinsic and
intrinsic motivation, social capital, social and personal cognition,
organization climate, and management championship (Alavi &
Leidner 1999; Bock & Kim, 2002; Bock et al., 2005; Chiu et al.,
2006; Hsu et al., 2007; Kankanhalli et al., 2005; Koh & Kim,
2004; Orlikowski 1996; Purvis et al., 2001; Wasko & Faraj, 2005).
Therefore, we could presume that individuals behavior for knowl-
edge sharing is affected by the contextual factors and personal per-
ceptions of the knowledge sharing in which they partake in. Social
cognitive theory (SCT) (Bandura, 1982, 1986, 1997) is a widely ac-
cepted model for validating individual behavior (Compeau & Hig-
gins, 1995). To investigate the knowledge sharing behavior in
PVCs, we use SCT to conceptualize a research model for this study.
In the SCT model, contextual factors, personal factors, and behavior
act as interacting relationships (Wood & Bandura, 1989). This
study focuses on the role of contextual factors and personal per-
ceptions on individual behavior.
The norm of reciprocity and trust are treated as two major con-
textual factors inuencing personal perceptions and a members
behavior. Knowledge sharing self-efcacy, perceived relative
advantage, and compatibility are seen as predictors of personal fac-
tors since they are all considered as the main inuences shaping
users behavior (Bandura, 1982, 1986, 1997; Igbaria & Iivari,
1995; Rogers, 2003; Sia, Teo, Tan, & Wei, 2004; Verhoef & Langerak,
2001).
2.2. Contextual factors
According to Davenport and Prusaks (1998) idea of a knowl-
edge market, the norm of reciprocity and trust are two of the most
signicant factors that drive knowledge sharing. Prior research
indicates that knowledge sharing in electronic networks is
facilitated by a strong sense of reciprocity favors given and re-
ceived along with a strong sense of fairness (Wasko & Faraj,
2000). Even though exchanges in electronic networks occur
through weak ties between strangers, there is evidence of recipro-
cal supportiveness (Wellman & Gulia, 1999). The norm of reciproc-
ity usually refers to a set of socially accepted rules regarding a
transaction in which a party extending a resource to another party
obligates the latter to return the favor (Wu et al., 2006), and has
been highlighted as a benet for individuals to engage in social ex-
change (Blau, 1964). A basic norm of reciprocity is a sense of mu-
tual indebtedness, so that individuals usually reciprocate the
benets they receive from others, ensuring ongoing supportive ex-
changes (Shumaker & Brownell, 1984). Different from altruism, the
norm of reciprocity represents a pattern of behavior where people
respond to friendly or hostile actions with similar actions (Fehr &
Gachter, 2000). This implies actions that are contingent on reward-
ing reactions from others and that cease when these expected reac-
tions are not forthcoming (Blau, 1964). The norm of reciprocity can
serve as a motivational mechanism for people to contribute to dis-
cretionary databases (Connolly & Thorn, 1990). Wasko and Faraj
 M.-J.J. Lin et al. / Computers in Human Behavior 25 (2009) 929939
(2000) suggested that people who share knowledge in online com-
munities believe in reciprocity.
Trust, an implicit set of beliefs that the other party will behave
in a dependent manner (Gefen, Karahanna, & Straub, 2003; Kumar,
Scheer, & Steenkamp, 1995) and will not take advantage of the sit-
uation (Gefen et al., 2003), has been also recognized as an impor-
tant factor affecting knowledge sharing (Ridings, Gefen, & Arinze,
2002) and as a determinant of the effectiveness of knowledge-
sharing activities (Chowdhury, 2005; Williams, 2001). Davenport
and Prusak (1998) indicated that trust must be visible; the mem-
bers of the organization must see people get credit for knowledge
sharing. They must directly experience reciprocity. According to
Blau (1964), trust creates and maintains exchange relationships,
which in turn may lead to sharing knowledge of good quality. Non-
aka (1994) indicated that inter-personal trust is important in
teams and organizations for creating an atmosphere for knowledge
sharing. Trust has been recognized as an important antecedent of
IS group performance and it has been found that knowledge shar-
ing is achieved through the mechanisms of mutual trust and inu-
ence between these groups (Nelson & Cooprider, 1996). Nahapiet
and Ghoshal (1998) suggested that when trust exists between
the parties, they are more willing to engage in cooperative
interaction.
2.3. Personal perceptions of knowledge sharing
Researchers noted that individual perceived attributes, such as
knowledge sharing self-efcacy, perceived relative advantage,
and compatibility, inuence member knowledge sharing in virtual
communities and organizations (Bock & Kim, 2002; Chiu et al.,
2006; Hsu et al., 2007; Kankanhalli et al., 2005; Wasko & Faraj,
2005).
Individuals pursuing improved perceptions of competency are
motivated by internal self-efcacy-based motivation. Self-efcacy
is a form of self-evaluation that inuences decisions about what
behaviors to undertake, the amount of effort and persistence to
put forth when faced with obstacles, and the mastery of the behav-
ior. In general, perceived self-efcacy plays an important role in
inuencing individuals motivation and behavior (Bandura, 1982,
1986; Igbaria & Iivari, 1995). For these individuals, the need for a
higher level of traits, competency, and values in their important
identities spurs them to perform a task. Therefore, people who
have high self-efcacy will be more likely to perform related
behavior than those with low self-efcacy (Schunk, 1990). This
construct has received considerable attention through empirical
research in the organizational behavior eld (Gist & Mitchell,
1992). More recently, the concept of self-efcacy has been applied
to knowledge management to validate the effect of personal ef-
cacy belief in knowledge sharing, i.e. knowledge sharing self-ef-
cacy. For example, in a motivation theory context, research has
found that knowledge sharing self-efcacy positively inuences
the motivation to share knowledge (Hsu et al., 2007; Kankanhalli
et al., 2005; Wasko & Faraj, 2005).
Relative advantage refers to the degree to which an innova-
tion provides more benets than its precursor. Relative advanta-
ges are manifested as increased efciency and effectiveness,
economic benets, and enhanced social status (Rogers, 2003).
Moore and Benbasat (1991) found that perceived relative advan-
tage of an innovation is positively related to the rate of adoption.
Correspondingly, facilitators of knowledge sharing reported obvi-
ous benets such as reduced communication costs and faster
problem-solving capability (Song, 2002). Therefore, when deci-
sion makers perceive clear overall personal and organizational
benets of knowledge sharing, they are more likely to facilitate
a knowledge-sharing culture in the organization (Kaser & Miles,
2002).
931
Individuals motivated by a purpose behave in congruity with
their value systems. Achieving internalized values and the purpose
of the new program or matter is the driving force behind this
source of motivation. Gerrard and Cunningham (2003) and Rogers
(2003) dened compatibility as the degree to which an innovation
is perceived as being consistent with existing values, previous
experiences, and potential needs, where existing values involve
lives style/habit, work attitude/relevance, and concepts in knowl-
edge sharing; previous experiences include using a computer, the
internet, new technology, and knowledge sharing in VCs; and po-
tential needs involve improved job performance, problem-solving
capability, innovation, and competitive advantage. Greater com-
patibility between personal policy and administrative innovation
is preferable because it is less uncertain to potential adopters
and permits innovation to be interpreted in a more familiar context
(Rogers, 2003).
3. Research model
Based on SCT, we may reasonably assume that the contexts of
PVCs shaped by the norm of reciprocity and trust should inuence
personal perceptions and a members behavior. Fig. 1 shows the re-
search model for this study, which examines the effects that the
norm of reciprocity has on knowledge sharing behavior through
trust and three personal perceptions. We also examine the rela-
tionship knowledge sharing behavior and community loyalty. Each
construct involved in the research model and hypotheses are dis-
cussed below.
In this study, the denition of the norm of reciprocity is peoples
salient belief that current knowledge sharing in VCs will lead to a
future request for knowledge being met (Davenport & Prusak,
1998). The norm of reciprocity refers to knowledge exchanges that
are mutual and perceived by the parties as obligatory and fair.
These arguments suggest a positive relationship between reciproc-
ity and the behavior of knowledge sharing by knowledge contrib-
utors. Knowledge sharing is encouraged and the tendency to
forego the temptation to free-ride suggests that the norm of reci-
procity governs online interaction (Constant, Sproull, & Kiesler,
1996; Rheingold, 2000; Wasko & Faraj, 2000; Wellman & Gulia,
1999). According to the social exchange theory (Thibaut & Kelley,
1959), the norm of reciprocity could underline the motivation
and commitment of community members to sharing knowledge
(Hall, 2001). When there is a strong norm of reciprocity in the col-
lective, knowledge contributors may feel obliged to share their
knowledge (Wasko & Faraj, 2005). In such a climate, knowledge
contributors are likely to share their knowledge with other mem-
bers in PVCs; the norm of reciprocity may be a salient motivator
for knowledge contributors. Thus, the following hypothesis is
proposed:
H1. The norm of reciprocity positively affects members knowl-
edge sharing behavior in PVCs.
Trust has been studied in a variety of online settings, and results
indicate that trust in others ability, benevolence, and integrity is
related to the desire to give and receive information (Ridings
et al., 2002) and to improved performance in distributed groups
(Jarvenpaa, 1998). In this study, the denition of trust is the degree
of belief in good intentions, behaviors, competence, and reliability
of members with respect to sharing knowledge in VCs. Different
from the norm of reciprocity, trust represents a promise keeping
of one party to the action of another party, irrespective of the abil-
ity to monitor or control the latter. Blau (1964) indicated that the
norm of reciprocity builds trust. Members must directly experience
reciprocity to establish trust (Davenport & Prusak, 1998). Further,
organizations with generalized reciprocity practices signal benevo-
lent intensions, which foster trust from members (Aselage & Eisen-
berger, 2003). Norms of reciprocity predisposes individuals to
932 M.-J.J. Lin et al. / Computers in Human Behavior 25 (2009) 929939
Fig. 1. Research model.
cooperate, understand, and empathize rather than to treat each
other as strangers, competitors, or potential enemies. In the early
days of a relationship, the norm of reciprocity is tacitly involved in
establishing social trust (Newton, 1997). As described above, trust
relationships can emerge within virtual communities without any
direct social interaction. This may be due to the transparency and
visibility of the norm of reciprocity and online knowledge-based
assets held in the communitys common repositories. Thus, when
there is a strong norm of reciprocity in the collective, individuals
trust that their knowledge contribution efforts will be recipro-
cated, thereby rewarding individual efforts and ensuring ongoing
contribution. This leads to the hypothesis:
H2. The norm of reciprocity positively affects members trust in
PVCs.
Trust has been dened as employees maintaining reciprocal
faith in each other in terms of intention and behaviors (Whitener,
2001). This study focuses on integrity, competence, reliability,
benevolence, and reciprocal faith, which refers to an individuals
expectation that members will follow a generally accepted set of
intentions, values, and benevolence in VCs. When employees have
high trust in employee relationships, they become more willing to
participate in knowledge-sharing activities (Abrams, Cross, Lesser,
& Levin, 2003; Lucas, 2005). Delahaye (2000) found that trust inu-
enced knowledge sharing decisions. Similarly, Corritore, Kracher,
and Wiedenbeck (2003) found trust to be a key element of success
in an on-line environment. Chowdhury (2005) demonstrated that
the presence of trust facilitates complex knowledge sharing.
Hence, the present study hypothesizes that trust positively inu-
ences personal behavior to facilitate knowledge sharing. The fol-
lowing hypothesis is thus proposed:
H3. Trust positively affects members knowledge sharing behav-
ior in PVCs.
As mentioned before, this study emphasizes the trust establish-
ment and treats trust as a contextual factor that is crucial to PVCs.
Comparing our research model with SCT, trust could be viewed as
an environmental factor and knowledge sharing self-efcacy could
be viewed as a personal factor. According to SCT, personal factors
are inuenced by environmental factors. Previous empirical re-
search has also found a causal relationship between trust and
self-efcacy (Cheung & Chan, 2000; Pavlou & Fygenson, 2006).
Cheung and Chan (2000) used SCT and other theories to examine
social cognitive factors in donating money to charity. Hsu et al.
(2007) used SCT theory to investigate the effect of identication-
based trust on knowledge sharing. Results show that there is a
strong and signicant direct effect of trust on self-efcacy; trust
raises the degree of self-efcacy for donation cognition. According
to Lave and Wengers (1991) theory of situated learning, a new
member in a virtual community becomes involved in a transition,
over time, from peripheral participation in the VC towards becom-
ing a masterful member. By sharing and developing ideas, by dis-
cussing, problem solving, and generally striving to become a
more competent member, the community members are able to en-
gage in the mutual development of both their own knowledge and
the communitys pool of expertise (Usoro et al., 2007). With this
ongoing process, members engaged in the development of cooper-
ative and trusting relationships while simultaneously developing
knowledge of what it means to be a competent and masterful
member (Nahapiet & Ghoshal, 1998). Therefore, the following
hypothesis is proposed:
H4. Trust positively affects members knowledge sharing self-
efcacy in PVCs.
Relative advantage is conceptualized as a multidimensional
construct that captures the benets of an innovation on lower
costs, savings in time and effort, and a decrease in discomfort (Rog-
ers, 2003). Handeld and Bechteln (2002) indicated that trust oc-
curs in cognitive and affect-based forms. Both forms were also
found to enhance coordination by lowering administrative costs.
Currall and Judge (1995) proposed that trust enhances the likeli-
hood of resource exchange between trusting parties, decreases
transaction costs by reducing the need for actions to protect the
interests of either party, decreases the costs of knowledge sharing,
and increases the likelihood that newly acquired knowledge can be
absorbed and retained. Dyer and Chu (2003) demonstrated that if
the suppliers trust the buyers, they are willing to share more
work-related information with the buyers. Tsai and Ghoshal
(1998) provided empirical evidence that suggests that trust inu-
ences resource exchange and combination. Prior studies have pro-
vided strong support for the signicant relationship between trust
and relative advantage.
As mentioned above, the denition of trust in this study focuses
on integrity, reliability and good intentions. Trust in the integrity of
a virtual community might be thought of as based in part on the
compatibility of the communitys cultural values with those of
the trusting member, the credibility of the communitys reputa-
tion, and the consistency of the communitys members past
behavior such as the extent to which actions are congruent with
 M.-J.J. Lin et al. / Computers in Human Behavior 25 (2009) 929939
words (Usoro et al., 2007). Consistent and reliable past behavior
create condence in future actions. If a member expects that other
members future behavior may lack integrity or unreliably, or is in
a manner that is otherwise incongruent with his personal values
and potential needs, he or she is unlikely to feel compatible. Con-
versely, the member is probably more willing to engage in cooper-
ative interaction when expectations of behavioral reliability and
perceptions of compatibility are high. These results imply a posi-
tive relationship between trust, potential advantages, and compat-
ibility. This leads to the hypotheses:
H5. Trust positively affects members perceived relative advan-
tage of knowledge sharing in PVCs.
H6. Trust positively affects members perceived compatibility of
knowledge sharing in PVCs.
Self-efcacy is an important concept in social psychology de-
rived from social cognitive theory. It can be dened as beliefs in
ones capabilities to organize and execute the courses of action re-
quired to produce given attainments (Jashapara & Tai, 2006). In
this study, knowledge sharing self-efcacy refers to the condence
in ones ability to provide knowledge that is valuable to others
(Kankanhalli et al., 2005). Researchers have found that individuals
with high levels of expertise, skills and capabilities are more likely
to provide useful advice on computer networks (Constant et al.,
1996). Through sharing useful knowledge to the organization, peo-
ple feel more condent in what they can do (Constant, Kiesler, &
Sproull, 1994). This perception of enhanced self-efcacy can moti-
vate employees to contribute their knowledge to others (Bock &
Kim, 2002). Hence, this study hypothesizes that knowledge sharing
self-efcacy positively inuences personal behavior to facilitate
knowledge-sharing activities as follows:
H7. Members knowledge sharing self-efcacy positively affects
their knowledge sharing behavior in PVCs.
In this study, perceived relative advantage refers to the knowl-
edge contributors cognition of likely advantages and benets that
the individuals knowledge sharing behavior will produce and
carry to him. Some studies (Andrews, 2002; Zhang & Hiltz, 2003)
suggested that individuals share knowledge within virtual commu-
nities to enrich their knowledge, to seek support, and to make
friends. Butler, Sproull, Kiesler, and Kraut (2002) suggested that
the primary reason for individuals to share knowledge is for their
benet in being seen as skilled, knowledgeable, or respected. Con-
sequently, a strong relative advantage of knowledge sharing per-
ceived by knowledge contributors positively inuences personal
behavior to facilitate knowledge-sharing activities. The following
hypothesis is thus proposed:
H8. Members perceived relative advantage positively affects
their knowledge sharing behavior in PVCs.
Perceived compatibility refers to the knowledge contributors
cognition of likely values, needs, and experience that the individ-
uals knowledge sharing behavior is in accordance with his or she
original value system. Budman (2003) considers perceived com-
patibility to be a psychological barrier, and believes that once indi-
viduals become comfortable with online transactions, they
increase the in usage of other online services. Moreover, a greater
t among components of knowledge sharing is desirable because it
can motivate individuals to develop new ideas (Hislop, 2003).
Thus, when members in VCs perceive knowledge sharing as com-
patible with their individual values and needs, they are more likely
to be positively predisposed to adopting and promoting it. The fol-
lowing hypothesis is proposed:
H9. Members perceived compatibility positively affects their
knowledge sharing behavior in PVCs.
Behaviors such as sharing knowledge (Abbott, 1988; Butler,
2001), disseminating ideas quickly (Finholt & Sproull, 1990), and
providing emotional support (King, 1994; Rice & Love, 1987) are
frequently observed in PVCs in the form of extensive postings
933
and viewings by members. Butler (2001) indicated that extensive
knowledge postings/viewings or frequent on-line interactions all
have the potential to support a higher level of help-giving behav-
iors and social support. Thus, when members frequently take part
in knowledge-sharing activities in PVCs, they are more likely to
positively promote PVCs or to invite new potential knowledge con-
tributors. The following hypothesis is proposed:
H10. Members knowledge sharing behavior positively affects
their community loyalty in PVCs.
4. Research methodology
We tested our hypotheses using data surveyed from three
maybe famous in Taiwan PVCs. We describe the development of
the survey instrument, the data collection procedures, and the val-
idation of the measures in the following subsections.
5.1. Survey procedure: Sample and data collection
An online survey was conducted because the target subjects
were individual members of three PVCs. The research model
was tested with data from members of Programmer-Club, Blue-
Shop, and Pure C in Taiwan. A banner with a hyperlink connecting
to our online survey was posted on the homepage of the Pro-
grammer-Club from April 20 to May 31, 2007, on the homepage
of the BlueShop from May1 to June 15, 2007, and on the home-
page of the Pure C from May 10 to June 30, 2007. Members with
knowledge sharing experience were invited to participate in this
survey. The three PVCs all provide several useful services and
mechanisms for members to share information and knowledge
with each other, such as le uploading for sharing, a technical for-
um, as electronic bulletin board, a special topic group, a chat
room, and e-mail services. 383 questionnaires were collected.
The exclusion of 33 invalid questionnaires resulted in a total of
350 complete and valid ones for data analysis. Table 1 lists the
demographic information of the respondents. To examine the rep-
resentative of the participating members, we performed MANOVA
to compare early respondents with late respondents on all of the
variables. The results suggest no signicant difference between
the early respondents and the late respondents (p < 0.05). Fur-
thermore, we conducted several diagnostic tests of common
methods bias effects; the results of these tests are reported in
the data analysis and results.
5.2. Measurement development
All the measurement items used in the study were adapted
from the literature, with minor modications to ensure contextual
consistency. Items used to operationally develop the constructs in-
cluded in each investigated model were mainly adapted from re-
lated theories and previous studies. Items of the norm of
reciprocity were developed and adapted from Wasko and Faraj
(2000), and Kankanhalli et al. (2005). Items of trust were adapted
from Lee and Choi (2003), and Ridings et al. (2002). Items of knowl-
edge sharing self-efcacy were based on Kalman (1999), Compeau
and Higgins (1995). Items of perceived relative advantage were
based on Moore and Benbasat (1991). Perceived compatibility
was assessed with items based on Moore and Benbasat (1991),
Nahapiet and Ghoshal (1998), and Tsai and Ghoshal (1998). Knowl-
edge sharing behavior was adapted and based on Davenport and
Prusak (1998). Community Loyalty was adapted and based on
Koh and Kim (2004), Srinivasan, Anderson, and Kishore (2002), Zei-
thaml, Berry, and Parasuraman (1996). For all the measures, a se-
ven-point Likert scale was adopted with anchors ranging from
strongly disagree (1), neither agree nor disagree(4), to strongly
agree (7).
934 M.-J.J. Lin et al. / Computers in Human Behavior 25 (2009) 929939

Table 1 6.1. Measurement model validation

Sample characteristics (the number of subjects = 350).

Demographic characteristics Frequency Percentage (%) We applied conrmatory factor analysis (CFA) (Bentler, 1992)
Gender Male 252 72 to evaluate the properties of the measures addressing latent con-
Female 98 28 structs. Values of skewness and kurtosis for all of the variables
Age <21 18 5 analyzed were well below prescribed levels (skewness < 2.0 and
2130 112 32 kurtosis < 7.0), indicating no signicant departures from normal-
3140 136 39 ity in the data (Yuan, Bentler, & Zhang, 2005). CFA indicated that
4150 53 15 the nal measurement model exhibited strong levels of t: v2/
>50 31 9
df = 1.555 (v2 = 325.09; df = 209), GFI = 0.93, AGFI = 0.90,
Education High school or below: 25 7 NFI = 0.96, NNFI = 0.98, CFI = 0.99, and RMSEA = 0.040, as shown
College (2 years): 67 19
University: 164 47
in Table 2. All the model-t indices exceed the respective com-
Graduate school 77 22 mon acceptance levels indicated by previous research, demon-
PhD 17 5 strating that the measurement model exhibited a fairly good t
Working experience <1 year 24 7 with the data collected.
13 32 9 Additionally, the convergent validity of the scales was veried
35 57 16 by using the three criteria suggested by Fornell and Larcker
510 123 35
(1981): (1) all indicator loadings should be signicant and exceed
1015 66 19
1520 28 8 0.7, (2) construct reliability should exceed 0.7(Hair et al., 1998),
>20 20 6 and (3) average variance extracted (AVE) by each construct should
Job Title IS manager: 98 28 exceed the variance due to measurement error for that construct
Project manager: 55 16 (i.e., AVE should exceed 0.50). In Table 3, it can be seen that the
Programmer: 24 7 CFA results were highly consistent with the relationships expected
Software engineer: 11 3 between the measured items and their respective constructs. All
Hardware engineer: 21 6
Web application engineer: 43 12
the factor loadings for all items exceed the recommended level of
System engineer: 27 8 0.7, and all factor loading are signicantly related, via t-tests at
Students: 28 8 p < 0.001, to their respective constructs. A series of Lagrange mul-
Others: 43 12 tiplier (LM) tests indicated that no items attempted to cross-load
Member history <3 month: 14 4 on a non-hypothesized construct. These analyses provided suf-
36 month: 21 6 cient evidence of unidimensionality. Composite reliability (CR)
6 month1 year: 42 12
was used to assess the internal consistency of the measurement
1 year2 year: 55 16
2 year3 year: 115 33 model. As shown in Table 3, the composite reliability of the con-
Over 3 year: 103 29 structs ranged from 0.82 to 0.91, and thus all exceeded the gener-
Online history <1 year: 13 4 ally accepted value of 0.70. In addition, the AVE ranged from 0.60
13 year: 31 9 to 0.76. Hence, all three conditions for convergent validity were
35 year: 71 20 met.
58 year: 118 34 Furthermore, the discriminant validity of the scales was as-
812 year: 77 22
Over 12 year: 40 11
sessed using the benchmark suggested by Fornell and Larcker
(1981): the square root of the AVE from the construct should be
greater than any of the inter-construct correlations. Table 4 lists
To establish content validity, the questionnaire was rened the correlations among the constructs, with the square root of
through cautious and rigorous pre-testing. A two-stage pretest of the AVE on the diagonal. All the diagonal values exceed the corre-
the questionnaire was performed by using 4 experts in the IS area lations between any pair of constructs, providing strong evidence
to assess instrument clarity, question wording, ease of understand- of discriminant validity at the construct level. Evidence of nomo-
ing and validity, logical consistencies, sequence of items, and con- logical validity was manifested in the inter-correlation matrix, as
textual relevance. The comments collected from these experts most of the correlations were in the expected direction and many
provided a basis for revisions to the construct measures and led expected associations were statistically signicant (see Table 4). In
to several minor modications of the wording and the item se- summary, the measurement model demonstrated adequate and
quence. Furthermore, a pilot study was conducted involving three sufcient reliability, convergent validity, discriminant validity,
professors, 15 Ph.D. candidates, and 20 masters degree candidates and nomological validity. Finally, we used Harmons single factor
who were members of various PVCs. Comments and suggestions
on the item contents and structure of the instrument were solic-
ited. The revised questionnaire items used to measure each con-
struct are listed in Appendix A. Table 2
Model t indices for measurement and structural models.

5. Data analysis and results Model t indices Measurement Structural Recommended

model model value
The research model shown in Fig. 1 was analyzed primarily Chi-square/degree of freedom 325.09/ 500.53/ 53.00
using SEM, supported by LISREL 8.7 software. Numerous research- (v2/d.f.) 209 = 1.555 220 = 2.275
ers have proposed a two-stage model-building process (Anderson Goodness-of-t index (GFI) 0.93 0.90 =0.90
Adjusted goodness-of-t index 0.90 0.86 =0.80
& Gerbing, 1988; Hair, Anderson, & Tatham, 1998; Joreskog &
(AGFI)
Sorbom, 2001; Maruyama, 1998). The rst step involves the analy- Normed t index (NFI) 0.96 0.95 =0.90
sis of the measurement model, while the second step tests the Non-normed t index (NNFI) 0.98 0.96 =0.90
structural relationships among latent constructs. The aim of Comparative t index (CFI) 0.99 0.97 =0.90
Root mean square error of 0.040 0.060 50.08
the two-step approach is to assess the reliability and validity of
approximation (RMSEA)
the measures before their use in the full model.
 M.-J.J. Lin et al. / Computers in Human Behavior 25 (2009) 929939 935

Table 3 test (McFarlin & Sweeney, 1992; Sanchez & Brock, 1996) to check
Conrmatory factor analysis results of measurement model. common method bias. The single-factor model resulted in
Construct and indicators Factor t-value Composite Average v2(230) = 4269.58, compared with v2(209) = 325.09 for the mea-
loadings reliability variance surement model, indicating that common method bias was not a
(CR) extracted serious threat in this study.
(AVE)
Norm of reciprocity (NR) 0.88 0.71
NR1 0.82 17.65 5.2. Test of the structural model
NR2 0.86 18.93
NR3 0.85 18.60
As shown in Table 2, all of the model-t indices of the struc-
Trust (TR) 0.90 0.75
tural model exceeded their respective common acceptance lev-
TR1 0.82 18.17
TR2 0.91 21.24 els: v2 to a degrees of freedom ratio of 2.275 (v2 = 500.53;
TR3 0.86 19.58 df = 220), GFI = 0.9, AGFI = 0.86, NFI = 0.95, NNFI = 0.96,
Knowledge sharing self-efcacy (KSSE) 0.91 0.76 CFI = 0.97, and RMSEA = 0.060, suggesting that the model t
KSSE1 0.88 20.25 the data well. The results of hypotheses tests along with the
KSSE2 0.89 20.74 path coefcients and their signicance values are shown in
KSSE3 0.85 19.15
Fig. 2. As expected, the norm of reciprocity was strong posi-
Perceived relative advantage (PRA) 0.82 0.60 tively related to trust (b = 0.36, p < 0.001), but it showed no sig-
PRA1 0.76 15.48
nicant inuence on knowledge sharing behavior. Consequently,
PRA2 0.80 16.50
PRA3 0.76 15.29 H2 was supported empirically while H1 was not. Trust is posi-
tively related to knowledge sharing behavior (b = 0.25, p < 0.01),
Perceived compatibility (PC) 0.89 0.67
PC1 0.85 19.10 knowledge sharing self-efcacy (b = 0.51, p < 0.001), perceived
PC2 0.85 19.03 relative advantage (b = 0.57, p < 0.001), and perceived compati-
PC3 0.78 16.78 bility (b = 0.32, p < 0.001). Hence, H3 to H6 were supported.
PC4 0.80 17.50 The results also revealed that knowledge sharing self-efcacy
Knowledge sharing behavior (KSB) 0.87 0.69 (b = 0.26, p < 0.001), perceived relative advantage (b = 0.31,
KSB1 0.83 18.39 p < 0.001), and perceived compatibility (b = 0.12, p < 0.05) were
KSB2 0.84 18.71
positively and signicantly related to knowledge sharing behav-
KSB3 0.82 17.95
ior, providing support for H7 to H9. H10 examined the relation-
Community loyalty (CL) 0.88 0.64
CL1 0.77 16.11
ship between knowledge sharing behavior and community
CL2 0.78 16.49 loyalty. The results show that knowledge sharing behavior had
CL3 0.82 17.66 a strong signicant inuence on community loyalty (b = 0.33,
CL4 0.82 17.94 p < 0.001), providing support for H10.

Table 4
Discriminant validity: Inter-Correlation and AVE.

Construct Mean St. Dev. AVE NR TR KSSE PRA PC KSB CL

NR 4.23 1.07 0.71 0.84
TR 5.29 1.14 0.75 0.36** 0.87
KSSE 5.10 1.19 0.76 0.19** 0.51** 0.87
PRA 4.73 1.09 0.60 0.21** 0.57** 0.29** 0.77
PC 4.16 0.97 0.67 0.12* 0.32** 0.16* 0.18* 0.82
KSB 5.15 1.06 0.69 0.21** 0.60** 0.50** 0.55** 0.30** 0.83
CL 4.60 0.91 0.64 0.07 0.19* 0.16* 0.18* 0.10* 0.33** 0.80

Note: Diagonal elements are the square root of the average variance extracted (AVE). Off-diagonal elements are the correlations among constructs. For discriminant validity,
AVE should be larger than squared correlation between any pair of constructs; hence diagonal elements should be larger than off-diagonal elements. p < 0.05; p < 0.01
(two-tailed test).

Fig. 2. SEM analysis of research model. *p < 0.05, **p < 0.01, ***
p < 0.001.
936 M.-J.J. Lin et al. / Computers in Human Behavior 25 (2009) 929939
6. Discussion and conclusion
6.1. Summary of results
This study developed an integrated model designed to investi-
gate and explain the relationships between contextual factors, per-
sonal perceptions of knowledge sharing, knowledge sharing
behavior, and community loyalty. Our results provide support for
the theoretical model and most of our hypotheses, and add to
the existing research on the validation of knowledge sharing
behavior on PVCs. This study produces four important ndings that
deserve considerable attention from executives of organizations
seeking to build a pool of rich innovation and knowledge, including
a comprehensive framework of the determinants to foster mem-
bers knowledge sharing behavior.
Firstly, the norm of reciprocity shows a negative and insigni-
cant inuence on knowledge sharing behavior. This nding directly
contradicts prior research in face-to-face settings, where it is con-
sistently found that reciprocity is critical for sustaining supportive
relationships and collective action (Putnam, 1995; Shumaker &
Brownell 1984). The result is similar to Wasko and Farajs (2005)
ndings that showed that the norm of reciprocity is not a signi-
cant predictor of helpfulness of knowledge contribution, but a sig-
nicant negative predictor of the volume of knowledge
contribution in electronic networks. One possible explanation is
that online-based interactions may be generalized rather than dya-
dic, and direct reciprocity is not necessary for sustaining collective
action (Wasko & Faraj, 2005). Generalized reciprocity occurs when
giving is not reciprocated by the recipient, but by a third party
(Ekeh, 1974). In contrast to personal knowledge sharing between
two individuals where there is an expectation of direct reciprocity,
reciprocity in PVCs may be generalized.
Secondly, according to Blau (1964), the norm of reciprocity
builds trust. This study provides supports that the norm of reci-
procity is a signicant determinant of trust in knowledge sharing.
That is, the norm of reciprocity is normative and supportive of
knowledge-sharing initiatives in PVCs. Members are more likely
to display condence and reliability in each others actions in rela-
tion to knowledge sharing. Hence, the norm of reciprocity exerts an
indirect positive effect on knowledge sharing behavior mediated
by trust. Tsai and Ghoshal (1998) found that social interaction
had a strong effect on trust in the context of resource exchange
and production innovation within an organization.
Thirdly, the results show that knowledge sharing self-efcacy,
perceived relative advantage, and perceived compatibility signi-
cantly and positively inuence knowledge sharing behavior. The
ndings reveal that members, who execute knowledge sharing to
be a capable, superior, and compatible means of achieving personal
objectives, express a high willingness to share their knowledge.
Moreover, as expected, trust enables knowledge sharing and has
a direct positive effect on knowledge sharing behavior. On the
other hand, trust exerts an indirect positive effect on knowledge
sharing behavior mediated by knowledge sharing self-efcacy, per-
ceived relative advantage, and perceived compatibility.
Lastly, knowledge sharing behavior may play a vital role under-
lying community loyalty in reaching a critical mass of community
members within a short time, and so brings plentiful knowledge,
potential members, and sustainable advantages to PVCs.
6.2. Implications for research
From a theoretical perspective, we have advanced our under-
standing of the contextual factors behind members knowledge
sharing behavior. Our rst nding implies that the norm of rec-
iprocity is an indirect but important inuencing factor for
knowledge sharing. The norm of reciprocity can motivate mem-
bers to share knowledge, but it is mediated by trust that leads to
a greater effect of knowledge sharing behavior. By identifying
facets of the norm of reciprocity and trust as the two key deter-
minants of knowledge sharing behavior, interactive relationships
among members of PVCs are characterized as a valuable context
for knowledge sharing. The norm of reciprocity falls along a con-
tinuum from negative reciprocity to generalized reciprocity
(Sahlins, 1972). If expectations of direct reciprocity are not crit-
ical to fostering knowledge sharing in PVCs, one potentially
interesting area of future research would be to apply multi-
group analysis techniques of SEM to examine whether general-
ized reciprocity is more important than direct reciprocity in
knowledge sharing.
Our second nding implies that the personal perceptions of
knowledge sharing (knowledge sharing self-efcacy, perceived rel-
ative advantage, and compatibility) as mediators of knowledge
sharing in professional virtual communities by themselves are suf-
cient for knowledge sharing behavior. Personal perceptions of
knowledge sharing can contribute to knowledge sharing behavior
to some extent. These ndings conform with and extend prior re-
search (Hsu et al., 2007; Kankanhalli et al., 2005).
Our study examined the effect of contextual factors and per-
sonal perceptions on knowledge sharing behavior and the relation-
ship between these factors. The results help explain knowledge
sharing behavior in PVCs. However, prior research suggests that a
greater level of knowledge sharing may lead to a better develop-
ment of contextual factors: interactive relationships, mutual trust,
identication, and shared vision. According to SCT (Bandura, 1997),
determinants inuence each other bidirectionally. Hence, such
relationships could be tested longitudinally. Future research
should look at changes in contextual factors and personal percep-
tions of knowledge sharing over time and the relationships of those
changes with knowledge sharing.
6.3. Implications for practitioners
This study proposes the following suggestions to help practi-
tioners manage or design better PVCs for facilitating members
knowledge sharing behavior. First of all, the results indicate that
the norm of reciprocity has an indirect effect through trust on
knowledge sharing behavior. From the practitioners standpoint,
the management of PVCs should foster a positive social interac-
tion culture before introducing knowledge-sharing initiatives.
Specically, creating a sharing climate characterized by manage-
ment support, members involvement, reciprocity, and reward sys-
tems in knowledge sharing and the knowledge community is
likely to facilitate both management and members to socialize
and interact frequently with one another, thus improving trust-
building and knowledge-sharing willingness. Managers of PVCs
can promote reciprocity by using incentives such as reputable re-
wards for sharing knowledge. For instance, a unique identity or
symbol can be provided to each member to motivate them to
contribute to build their own reputations. The BlueShop commu-
nity web has a creative mechanism by which knowledge receivers
can donate value-added points (VP) knowledge contributors as a
reward for favors. Acquiring value-added points through contrib-
uting knowledge is a visible reputation symbol that motivates
knowledge sharing and minimizes free-riders. It also positively
encourages a member to reciprocate the benets that he or she
received from other members. The score of value-added points
may represent knowledge contributors professional status, credit,
or reputation within the community. Members feel an immediate
reward from their contribution and hence increase their sharing
(Tiwana & Bush, 2001).
Secondly, the results indicate that trust has both direct and
indirect effects on knowledge sharing behavior, implying that trust
 M.-J.J. Lin et al. / Computers in Human Behavior 25 (2009) 929939
plays a critical role in promoting the behavior of knowledge shar-
ing within PVCs. Managers must consider that knowledge sharing
can only occur when members are willing to share their knowl-
edge. Trust helps eliminate barriers to knowledge sharing. Without
trust, the cooperation required for successful knowledge sharing
might not exist. Hence, when promoting knowledge sharing, man-
agers of PVCs should develop solid mechanisms that improve the
interactive quality and ties of the relationships among members.
For example, the Programmer-Club community web often holds
face-to-face seminars, and invites knowledgeable contributors
and professional instructors to share their knowledge and experi-
ence with members of the community, as a good way of enhancing
the social interaction among its members. The Programmer-Club,
BlueShop, and Pure C communities also provide personal message
boards, chat rooms, technical forums, and blogs as tools for
improving online communication and contacts among members.
In addition, managers can improve the trust relationships among
members by facilitating the norm of reciprocity, sharing vision
(experiences), dialoguing, and by conding personal information
in PVCs. For example, Programmer-Club provides a convenient ser-
vice for members to disclose personal information when register-
ing as a new member and allows the browsing of other
members messages to get more acquainted with them.
Lastly, since knowledge sharing self-efcacy, perceived relative
advantage, and compatibility are important predictors of motiva-
tion to facilitate knowledge sharing, management of PVCs should
provide some facilities, such as on-line training programs, support
mechanisms, and guidelines, to increase members self-efcacy so
that members are condent enough to share their knowledge in
PVCs. Furthermore, the management of PVCs should actively
search for useful information on the benets of knowledge shar-
ing. For instance, Programmer-Club and BlueShop are successful
professional online communities. Their main strategy is to be-
come members of technical alliance programs, receive online
advertising, and win awards for excellent virtual communities
to enhance their reputation and to meet members expectations
and needs.
6.4. Limitations and future research
Although our ndings are useful, there are several limitations to
this study, requiring further examination and additional research.
First, the results should be interpreted as only explaining knowl-
edge sharing of current knowledge contributors of VCs. Whether
the results can be generalized to non-participants or to disaffected
participants will require additional research. The sample was col-
lected from Taiwanese PVCs; the research model should be tested
further using samples from other countries, since cultural differ-
ences among VCs inuence member perception of sharing knowl-
edge. Further testing would provide a more robust test of the
hypotheses. Second, based on a sample of 350 respondents from
three PVCs, it is unclear whether our ndings and research model
could be generalized to all types of professional virtual communi-
ties. The study ndings should be veried with more types of VC
and a larger sample to increase its generalizability. Further studies
can examine other determinants, such as different kinds of knowl-
edge sharer (heavy sharer, medium sharer, light sharer, and free-ri-
der). Different kinds of knowledge sharer may have different
inuences on the knowledge sharing process.
Appendix A. Scales and measurement items used in the study
Please use the following scale to indicate your extent of agree-
ment about how well each of the following statements is an accu-
rate description of your virtual community. Here, 1 = strongly
disagree, 7 = strongly agree.
937
A.1. Norm of reciprocity (NR)
Denition: Peoples salient beliefs that current knowledge shar-
ing to VCs would lead to future request for knowledge being met
(Davenport & Prusak, 1998).
I know that other members will help me, so its obligator and fair
to help other members in this virtual community.
When I share knowledge with other members, I believe that the
members in this virtual community would help me if I need it.
When I share knowledge with other members, I believe that my
queries for knowledge will be answered in the future in this vir-
tual community.
A.2. Trust (TR)
Denition: The degree of belief in good intentions, behaviors,
competence, and reliability of members with respect to sharing
knowledge in VCs (Lee & Choi, 2003; Mishra, 1996).
Members in this virtual community have reciprocal faith-based
and trustworthy relationships.
Members in this virtual community will not take advantage of
others even when a protable opportunity arises.
Members in this virtual community always keep promises that
they make to one another.
Please use the following scale to indicate your extent of agree-
ment about how well each of the following statements is an accu-
rate description of your perceptions in virtual community. Here,
1 = strongly disagree, 7 = strongly agree.
A.3. Knowledge sharing self-efcacy (KSSE)
Denition: The degree of condence in members ability to shar-
ing knowledge that is valuable to the VCs (Constant et al., 1996;
Kalman, 1999).
I have condence in my ability to provide knowledge that other
members in this virtual community consider valuable.
I have the expertise, experiences, and insights needed to provide
knowledge that is valuable for other members in this virtual
community.
I have condence in responding or adding comments to mes-
sages or articles posted by other members in this virtual
community.
A.4. Perceived relative advantage (PRA)
Denition: The degree to which encouraging knowledge sharing
is perceived to benet the conduct of members (Moore & Benbasat,
1991).
Sharing knowledge with members in this virtual community
will increase my solving-problem capability.
Sharing knowledge with members in this virtual community
will rapidly absorb and react to new information regarding the
area.
Sharing knowledge with members in this virtual community
will help me in my job and improve my performance.
A.5. Perceived compatibility (PC)
Denition: The degree to which encouraging knowledge sharing
is perceived to t the value system and current needs of members
(Moore & Benbasat, 1991).
938 M.-J.J. Lin et al. / Computers in Human Behavior 25 (2009) 929939
Sharing knowledge with members in this virtual community is
compatible with my values.
Sharing knowledge with members in this virtual community ts
my current needs.
Sharing knowledge with members in this virtual community is
compatible with my previous experiences.
Sharing knowledge with members in this virtual community ts
my work style.
Please use the following scale to indicate your extent of agree-
ment about how well each of the following statements is an accu-
rate description of your behavior. Here, 1 = strongly disagree,
7 = strongly agree.
A.6. Knowledge sharing behavior (KSB)
Denition: The degree to which a member conducts knowledge-
sharing activities in VCs (Davenport & Prusak, 1998).
I frequently participate in knowledge-sharing activities and
share my knowledge with others in this virtual community.
I usually spend a lot of time conducting knowledge-sharing
activities in this virtual community.
When discussing a complicated issue, I am usually involved in
the subsequent interactions.
A.7. Community loyalty (CL)
Denition: The degree to which a member promotes the virtual
community to get new members to join and talks about the bene-
ts of this VC (Srinivasan et al., 2002; Zeithaml et al., 1988).
I frequently talk to people about the benets of our virtual
community.
I usually spend some time providing useful suggestions for our
virtual community.
I often introduce my peers or friends to our virtual community.
I actively invite my close acquaintances to join our virtual
community.
References
Abbott, A. (1998). The system of professions: An essay on the division of expert labor.
Chicago, IL: University of Chicago Press.
Abrams, L. C., Cross, R., Lesser, E., & Levin, D. Z. (2003). Nurturing interpersonal trust
in knowledge-sharing networks. Academy of Management Executive, 17(4),
6477.
Alavi, M., & Leidner, D. E. (1999). Knowledge management systems: Issues,
challenges and benets. Communication of Association for Information Systems,
1, 128.
Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice. A
review and recommended two-step approach. Psychological Bulletin, 103,
411423.
Andrews, D. (2002). Audience-specic online community design. Communications of
the ACM, 45(4), 6468.
Argote, L., Ingram, P., Levin, J. M., & Moreland, R. L. (2000). Organizational learning:
Creating, retaining, and transferring knowledge. Boston, MA: Kluwer Academic
Publishers.
Aselage, J., & Eisenberger, R. (2003). Perceived organizational support and
psychological contracts: A theoretical integration. Journal of Organizational
Behavior, 24, 491509.
Bakker, M., Leenders, R. T. A. J., Garray, S. M., Kratzer, J., & Van Engelen, J. M. L.
(2006). Is trust really social capital? Knowledge sharing in product
development projects. The Learning Organization, 13(6), 594607.
Bandura, A. (1982). Self-efcacy mechanism in human agency. American
Psychologist, 37(2), 122147.
Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory.
Englewood Cliffs, NJ: Prentice-Hall.
Bandura, A. (1997). Self-efcacy: The exercise of control. New York: W.H. Freeman
and Company.
Bentler, P. M. (1992). EQS structural equations program manual. Los Angeles:
Statistical Software, Inc..
Blau, P. M. (1964). Exchange and power in social life. New York: Wiley.
Bock, G. W., & Kim, Y. G. (2002). Breaking the myths of rewards: An exploratory
study of attitudes about knowledge sharing. Information Resource Management
Journal, 15(2), 1421.
Bock, G. W., Zmud, R. W., Kim, Y. G., & Lee, J. N. (2005). Behavioral intention
information in knowledge sharing: Examining the roles of extrinsic motivators,
social-psychological forces, and organizational climate. MIS Quarterly, 29(1),
87111.
Budman, M. (2003). Internet life: What your customers are doing online. Across the
Board, 40(1), 5960.
Butler, B. S. (2001). Membership size, community activity, and sustainability: A
resource-based model of online social structures. Information Systems Research,
12(4), 346362.
Butler, B., Sproull, L., Kiesler, S., & Kraut, R. (2002). Community effort in online
groups: Who does the work and why. In S. Weisband & L. Atwater (Eds.),
Leadership at a distance. Mahwah, NJ: Lawrence Erlbaum Publishers.
Cheung, C. K., & Chan, C. M. (2000). Social-cognitive factors of donating money to
charity, with special attention to an international relief organization. Evaluation
and Program Planning, 23(2), 241253.
Chiu, C. M., Hsu, M. H., & Wang, Eric T. G. (2006). Understanding knowledge sharing
in virtual communities: An integration of social capital and social cognitive
theories. Decision Support Systems, 42, 18721888.
Chowdhury, S. (2005). The role of affect- and cognition-based trust in complex
knowledge sharing. Journal of Managerial Issues, 17(3), 310326.
Compeau, D. R., & Higgins, C. A. (1995). Computer self-efcacy development of a
measure and initial test. MIS Quarterly, 19(2), 189211.
Connolly, T., & Thorn, B. K. (1990). Discretionary databases: Theory, data, and
implications. In J. Fulk & C. Steineld (Eds.), Organizations and communication
technology (pp. 219233). Newbury Park, CA: Sage Publications.
Constant, D., Kiesler, S., & Sproull, L. (1994). Whats mine is ours, or is it? A study of
attitudes about information sharing. Information Systems Research, 5(4),
400421.
Constant, D., Sproull, L., & Kiesler, S. (1996). The kindness of strangers: The
usefulness of electronic weak ties for technical advice. Organization Science,
7(2), 119135.
Corritore, C. L., Kracher, B., & Wiedenbeck, S. (2003). On-line trust: Concepts,
evolving themes, a model. International Journal of HumanComputer Studies,
58(6), 737758.
Corso, M., Martini, A., Paolucci, E., & Pellegrini, L. (2003). Technological and
organizational tools for knowledge management: In search for congurations.
Small Business Economics, 21(4), 397408.
Currall, S., & Judge, T. (1995). Measuring trust between organizational boundary
role persons. Organizational Behavior and Human Decision Processes, 64(2),
151170.
Davenport, T. H., & Prusak, L. (1998). Working knowledge: How organizations manage
what they know. Boston: Harvard Business School Press.
Delahaye, B. L. (2000). Human resource development: Principles and practice.
Brisbane: John Wiley & Sons (Australia).
Dyer, J. H., & Chu, W. (2003). The role of trustworthiness in reducing transaction
costs and improving performance. empirical evidence from the United States,
Japan, and Korea. Organization Science, 14(1), 5768.
Ekeh, P. P. (1974). Social exchange theory: The two traditions. Cambridge, MA:
Harvard University Press.
Fehr, E., & Gachter, S. (2000). Fairness and retaliation: The economics of reciprocity.
Journal of Economic Perspectives, 14(40), 159181.
Fernback, J. (1999). There is a there: Notes toward a denition of cyber-community.
In S. G. Jones (Ed.), Doing internet research: Critical issues and methods for examine
the net (pp. 203220). Thousand Oaks, CA: Sage Publications.
Finholt, T., & Sproull, L. (1990). Electronic groups at work. Organization Science, 1,
4164.
Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with
unobservable variables and measurement error. Journal of Marketing Research,
18(1), 3950.
Gefen, D., Karahanna, E., & Straub, D. W. (2003). Trust and TAM in online shopping:
An integrated model. MIS Quarterly, 27(1), 5190.
Gerrard, P., & Cunningham, J. B. (2003). The diffusion of internet banking
among Singapore consumers. International Journal of Bank Marketing, 21(1),
1628.
Gist, M. E., & Mitchell, T. R. (1992). Self-efcacy a theoretical-analysis of its
determinants and malleability. Academy of Management Review, 17(2),
183211.
Gongla, P., & Rizzuto, C. R. (2001). Evolving communities of practice. IBM global
services experience. IBM Systems Journal, 40(4), 842862.
Hair, J. F., Anderson, R. L., & Tatham, W. C. (1998). Multivariate data analysis with
reading. Upper Saddle River, NJ: Prentice-Hall.
Hall, H. (2001). Social Exchanges for knowledge exchange paper presented at
managing knowledge: Conversations and critiques. University of Leicester
Management Centre, 1011. Available from: http://www.bim.napier.ac.uk/esis/
about_us/hazel_publications.html.
Handeld, R. B., & Bechtelb, C. (2002). The role of trust and relationship structure in
improving supply chain responsiveness. Industrial Marketing Management, 31,
367382.
Hiltz, S. R., & Wellman, B. (1997). Asynchronous learning networks as a virtual
classroom. Communications of ACM, 40(9), 4449.
 M.-J.J. Lin et al. / Computers in Human Behavior 25 (2009) 929939
Hsu, M. H., Ju, T. L., Yen, C. H., & Chang, C. M. (2007). Knowledge sharing behavior in
virtual communities: The relationship between trust, self-efcacy, and outcome
expectations. International Journal of HumanComputer Studies, 65, 153169.
Hislop, D. (2003). Linking human resource management and knowledge
management via commitment. Employee Relations, 25(2), 182202.
Igbaria, M., & Iivari, J. (1995). The effects of self-efcacy on computer usage. Omega,
23(6), 587605.
Jarvenpaa, S. (1998). Is anybody out there? Antecedents of trust in global virtual
teams. Journal of Management Information Systems, 14(4), 2965.
Joreskog, K., & Sorbom, D. (2001). LISREL 8.7: Users Reference Guide. Chicago, IL:
Scientic Software International.
Jashapara, A., & Tai, W. C. (2006). Understanding the complexity of human
characteristics on e-learning systems: An integrated study of dynamic
individual differences on user perceptions of ease of use. Knowledge
Management Research & Practice, 4, 227239.
Kalman, M. E. (1999). The Effects of organizational commitment and expected
outcomes on the motivation to share discretionary information in a
collaborative database: Communication dilemmas and other serious games.
Unpublished Ph.D. Thesis, University of Southern California, Los Angeles, CA.
Kankanhalli, A., Tan, B. C. Y., & Wei, K. K. (2005). Contributing knowledge to
electronic knowledge repositories: An empirical investigation. MIS Quarterly,
29(1), 113143.
Kaser, P. A. W., & Miles, R. E. (2002). Understanding knowledge activists successes
and failures. Long Range Planning, 35(1), 928.
King, S. (1994). Analysis of electronic support groups for recovering addicts.
Interpersonal Computer Technology, 2(3), 4756.
Koh, J., & Kim, Y. G. (2004). Knowledge sharing in virtual communities: An e-
business perspective. Expert Systems with Applications, 26(2), 155166.
Kumar, N., Scheer, L. K., & Steenkamp, J. B. E. M. (1995). The effects of perceived
interdependence on dealer attitudes. Journal of Marketing Research, 17, 348356.
Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation.
Cambridge, MA: Cambridge University Press.
Lee, H., & Choi, B. (2003). Knowledge management enablers, processes, and
organizational performance. An integrative view and empirical examination.
Journal of Management Information System, 20(1), 179228.
Lucas, L. M. (2005). The impact of trust and reputation on the transfer of best
practices. Journal of Knowledge Management, 9(4), 87101.
Maruyama, G. M. (1998). Basic of structural equation modeling. Thousand Oaks: Sage
Publications.
McFarlin, D. B., & Sweeney, P. D. (1992). Distributive and procedural justice as
predictors of satisfaction with personal and organizational outcomes. Academy
of Management Journal, 35(3), 626.
Mishra, A. K. (1996). Organizational responses to crisis: The centrality of trust. In R.
M. Kramer & T. R. Tyler (Eds.), Trust in organizations: Frontiers of theory and
research (pp. 261287). Thousand Oaks, CA: Sage Publications.
Moore, G. C., & Benbasat, I. (1991). Development of an instrument to measure the
perceptions of adopting an information technology innovation. Information
Systems Research, 2(3), 173191.
Nahapiet, J., & Ghoshal, S. (1998). Social capital, intellectual capital, and
organizational advantage. Academy of Management Review, 23(2), 242266.
Nambisan, S., & Sawhney, M. (2007). The global brain: Your roadmap for innovating
faster and smarter in a networked world (1st ed.). Wharton School Publishing.
Nelson, K. M., & Cooprider, J. G. (1996). The contribution of shared knowledge to IS
group performance. MIS Quarterly, 21(4), 409429.
Newton, K. (1997). Social capital and democracy. American Behavioral Scientist,
40(5), 575589.
Nonaka, I. (1991). The knowledge creating company. Harvard Business Review,
NovemberDecember (pp. 96104).
Nonaka, I. (1994). A dynamic theory of organizational knowledge creation.
Organization Science, 5(1), 1435.
Orlikowski, W. J. (1996). Learning from notes: Organizational issues in groupware
implementation. In R. Kling (Ed.), Computerization and controversy
(pp. 173189). New York: Academic Press.
Pavlou, P. A., & Fygenson, M. (2006). Understanding and predicting electronic
commerce adoption: An extension of the theory of planned behavior. MIS
Quarterly, 30(1), 115143.
Petrash, G. (1996). Dows journey to a knowledge value management culture.
European Management Journal, 14, 365373.
Pfeffer, J., & Sutton, R. (1999). Knowledge What to do is not enough: Turning
knowledge into action. California Management Review, 42(1), 83108.
Preece, J. (2000). Online communities: Designing usability, supporting sociability. New
York: Wiley.
Purvis, R. L., Sambamurthy, V., & Zmud, R. W. (2001). The assimilation of knowledge
platforms in organizations: An empirical investigation. Organization Science,
12(2), 117135.
939
Putnam, R. (1995). Tuning in, tuning out: The strange disappearance of social capital
in America. Political Science and Politics, 66, 4683.
Rheingold, H. (2000). The virtual community: Homesteading on the electronic frontier.
Cambridge, MA: MIT Press.
Rice, R. E., & Love, G. (1987). Electronic emotion: Socioemotional content in a
computer-mediated communication network. Communication Research, 14(1),
85108.
Ridings, C. M., Gefen, D., & Arinze, B. (2002). Some antecedents and effects of
trust in virtual communities. Journal of Strategic Information Systems, 11,
271295.
Rogers, E. M. (2003). Diffusion of innovations (5th ed.). New York: Free Press.
Rothaermel, F. T., & Sugiyama, S. (2001). Virtual Internet communities and
commercial success: Individual and community-level theory grounded in the
atypical case of TimeZone.com. Journal of Management, 27(3), 297312.
Ryu, S., Ho, S. H., & Han, I. (2003). Knowledge sharing behavior of physicians in
hospitals. Expert Systems with Applications, 25(1), 113122.
Sahlins, M. (1972). Stone age economics. New York: Aldine de Gruyter.
Sanchez, J. I., & Brock, P. (1996). Outcomes of perceived discrimination among his
panic employees: Is diversity management a luxury or a necessity? Academy of
Management Journal, 39(3), 704.
Schunk, D. H. (1990). Goal setting and self-efcacy during self-regulated learning.
Educational Psychologist, 25(1), 7186.
Shumaker, S., & Brownell, A. (1984). Toward a theory of social support: Closing
conceptual gaps. Journal of Social Issues, 40(4), 1136.
Sia, C. L., Teo, H. H., Tan, B. C. Y., & Wei, K. K. (2004). Effects of environmental
uncertainty on organizational intention to adopt distributed work
arrangements. IEEE Transaction on Engineering Management, 51(3), 253267.
Song, S. (2002). An internet knowledge sharing systems. Journal of Computer
Information System, 42(3), 2530.
Srinivasan, S. S., Anderson, R., & Kishore, P. (2002). Customer loyalty in e-commerce.
An exploration of its antecedents and consequences. Journal of Retailing, 78(1),
4150.
Szulanski, G. (1996). Exploring internal stickiness: Impediments to the transfer of
best practices within the rm. Strategic Management Journal, 17, 2743.
Thibaut, J. W., & Kelley, H. H. (1959). The social psychology of groups. New York: John
Wiley.
Tiwana, A., & Bush, A. (2001). A social exchange architecture for distributed web
communities. Journal of Knowledge Management, 5(3), 242248.
Tsai, W., & Ghoshal, S. (1998). Social capital and value creation: The role of intrarm
networks. Academy of Management Journal, 41, 464478.
Usoro, A., Sharratt, M. W., Tsui, E., & Shekhar, S. (2007). Trust as an antecedent to
knowledge sharing in virtual communities of practice. Knowledge Management
Research & Practice, 5(3), 199212.
Verhoef, P. C., & Langerak, F. (2001). Possible determinants of consumers adoption
of electronic grocery shopping in the Netherlands. Journal of Retailing and
Consumer Services, 8(5), 275285.
Wasko, M. M., & Faraj, S. (2000). It is what one does: Why people participate and
help others in electronic communities of practice. Journal of Strategic
Information Systems, 9(23), 155173.
Wasko, M. M., & Faraj, S. (2005). Why should I share? Examining social capital and
knowledge contribution in electronic networks of practice. MIS Quarterly, 29(1),
3558.
Wellman, B., & Gulia, M. (1999). Virtual communities as communities: Net surfers
dont ride alone. In M. A. Smith & P. Kollock (Eds.), Communities in cyberspace
(pp. 167194). London: Routledge.
Whitener, E. M. (2001). Do high commitment human resource practices affect
employee commitment? A cross-level analysis using hierarchical linear
modeling. Journal of Management, 27(5), 515535.
Wiig, K. M. (1997). Knowledge management: An introduction and perspective. The
Journal of Knowledge Management, 1(1), 614.
Williams, M. (2001). In whom we trust: Group membership as an affective context
for trust development. Academy of Management Review, 26(3), 377397.
Wood, R., & Bandura, A. (1989). Social cognitive theory of organizational
management. Academy of management Review, 14(3), 361384.
Wu, J. B., Hom, P. W., Tetrick, L. E., Shore, L. M., Jia, L., Li, C., et al. (2006). The norm of
reciprocity: Scale development and validation in the Chinese context.
Management and Organization Review, 2(3), 377402.
Yuan, K., Bentler, P. M., & Zhang, W. (2005). The effect of Skewness and Kurtosis on
mean and covariance structure analysis. Sociological Methods and Research,
34(2), 240258.
Zeithaml, V. A., Berry, L. L., & Parasuraman, A. (1996). The behavioral consequences
of service quality. Journal of Marketing, 60(2), 3146.
Zhang, Y., & Hiltz, S. R. (2003). Factors that inuence online relationship
development in a knowledge sharing community. In Proceedings of the ninth
American conference on information systems (pp. 410417).